Proteins of glutamatergic NMDA receptor signaling pathways have been studied as targets for intervention in a variety of neuropathological conditions, including neurodegenerations, epilepsy, neuropathic pain, drug addiction, and schizophrenia. High activity NMDA-blocking agents have been designed to treat some of these disorders; however, their effect is often compromised by undesirable side effects. Therefore, alternative ways of modulating NMDA receptor function need to be sought after. The opening of the NMDA receptor ion channel requires occupation of two distinct binding sites, the glutamate site and the glycine site. It has been shown that D-serine, rather than glycine, can trigger the physiological NMDA receptor function. D-serine is a product of the activity of a specific enzyme, serine racemase (SR), which was identified a decade ago. SR has therefore emerged as a new potential target for the NMDA-receptor-based diseases. There is evidence linking increased levels of D-Ser to amyotrophic lateral sclerosis and Alzheimer's disease and decreased concentrations of D-serine to schizophrenia. SR is a pyridoxal-5'-phosphate dependent enzyme found in the cytosol of glial and neuronal cells. It is activated by ATP, divalent cations like Mg(2+) or Ca(2+), and reducing agents. This paper reviews the present literature on the activity and inhibition of mammalian SRs. It summarizes approaches that have been applied to design SR inhibitors and lists the known active compounds. Based on biochemical and docking analyses, i) we delineate for the first time the ATP binding site of human SR, and ii) we suggest possible mechanisms of action for the active compounds. In the end, we discuss the SR features that make the discovery of its inhibitors a challenging, yet very important, task of medicinal chemistry.